cgroup,writeback: don't switch wbs immediately on dead wbs if the memcg is dead

cgroup writeback tries to refresh the associated wb immediately if the
current wb is dead.  This is to avoid keeping issuing IOs on the stale
wb after memcg - blkcg association has changed (ie. when blkcg got
disabled / enabled higher up in the hierarchy).

Unfortunately, the logic gets triggered spuriously on inodes which are
associated with dead cgroups.  When the logic is triggered on dead
cgroups, the attempt fails only after doing quite a bit of work
allocating and initializing a new wb.

While bab3d59e196c ("mm/filemap.c: don't initiate writeback if mapping
has no dirty pages") alleviated the issue significantly as it now only
triggers when the inode has dirty pages.  However, the condition can
still be triggered before the inode is switched to a different cgroup
and the logic simply doesn't make sense.

Skip the immediate switching if the associated memcg is dying.

This is a simplified version of the following two patches:

 * https://lore.kernel.org/linux-mm/20190513183053.GA73423@dennisz-mbp/
 * http://lkml.kernel.org/r/156355839560.2063.5265687291430814589.stgit@buzz

Cc: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Fixes: 63d2cb77a4c6 ("writeback: disassociate inodes from dying bdi_writebacks")
Acked-by: Dennis Zhou <dennis@kernel.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 8461a63220398024c2aef5d4795d2f2aadd4ebfa..335607b8c5c001ca35847c5ec22409ab04d83b05 100644 (file)
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -576,10 +576,13 @@ void wbc_attach_and_unlock_inode(struct writeback_control *wbc,
        spin_unlock(&inode->i_lock);
 
        /*
-        * A dying wb indicates that the memcg-blkcg mapping has changed
-        * and a new wb is already serving the memcg.  Switch immediately.
+        * A dying wb indicates that either the blkcg associated with the
+        * memcg changed or the associated memcg is dying.  In the first
+        * case, a replacement wb should already be available and we should
+        * refresh the wb immediately.  In the second case, trying to
+        * refresh will keep failing.
         */
-       if (unlikely(wb_dying(wbc->wb)))
+       if (unlikely(wb_dying(wbc->wb) && !css_is_dying(wbc->wb->memcg_css)))
                inode_switch_wbs(inode, wbc->wb_id);
 }
 EXPORT_SYMBOL_GPL(wbc_attach_and_unlock_inode);